Trust Assessment
pptx-creator received a trust score of 10/100, placing it in the Untrusted category. This skill has significant security findings that require attention before use in production.
SkillShield's automated analysis identified 10 findings: 1 critical, 3 high, 5 medium, and 1 low severity. Key findings include Arbitrary command execution, Unsafe deserialization / dynamic eval, Dangerous call: subprocess.run().
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. The Manifest Analysis layer scored lowest at 42/100, indicating areas for improvement.
Last analyzed on February 14, 2026 (commit 13146e6a). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings10
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| CRITICAL | Arbitrary command execution Python shell execution (os.system, subprocess) Review all shell execution calls. Ensure commands are static (not built from user input), use absolute paths, and are strictly necessary. Prefer library APIs over shell commands. | Manifest | skills/dbhurley/pptx-creator/scripts/create_pptx.py:204 | |
| HIGH | Dangerous call: subprocess.run() Call to 'subprocess.run()' detected in function 'generate_image'. This can execute arbitrary code. Avoid using dangerous functions like exec/eval/os.system. Use safer alternatives. | Static | skills/dbhurley/pptx-creator/scripts/create_pptx.py:204 | |
| HIGH | Arbitrary File Read via Path Traversal Multiple scripts allow reading arbitrary files from the filesystem due to unsanitized user-provided paths. An attacker can supply a path like `../../../../etc/passwd` to arguments such as `--outline`, `--json`, `--template`, `--from`, or image paths specified within outline/JSON data. The content of these files would then be processed by the skill, potentially leading to data exfiltration through the skill's output (e.g., `analyze_template.py` printing file content, or `create_pptx.py` embedding it in a presentation). Sanitize all user-provided file paths to ensure they are within expected, designated directories (e.g., using `pathlib.Path.resolve()` and checking against a base directory, or disallowing `..` in paths). Implement strict validation for file types and content. | LLM | scripts/create_pptx.py:100 | |
| HIGH | Arbitrary File Write via Path Traversal Multiple scripts allow writing files to arbitrary locations on the filesystem due to unsanitized user-provided output paths. An attacker can supply a path like `../../../../tmp/malicious.pptx` to arguments such as `--output` or `--save-template` (for the template name part). This could overwrite critical system files or place malicious files in sensitive locations. Sanitize all user-provided output paths to ensure they are within designated output directories. Disallow `..` in paths and use `pathlib.Path.resolve()` to prevent path traversal. | LLM | scripts/create_pptx.py:430 | |
| MEDIUM | Unsafe deserialization / dynamic eval Decryption followed by code execution Remove obfuscated code execution patterns. Legitimate code does not need base64-encoded payloads executed via eval, encrypted-then-executed blobs, or dynamic attribute resolution to call system functions. | Manifest | skills/dbhurley/pptx-creator/scripts/create_pptx.py:189 | |
| MEDIUM | Unsafe deserialization / dynamic eval Decryption followed by code execution Remove obfuscated code execution patterns. Legitimate code does not need base64-encoded payloads executed via eval, encrypted-then-executed blobs, or dynamic attribute resolution to call system functions. | Manifest | skills/dbhurley/pptx-creator/scripts/use_template.py:8 | |
| MEDIUM | Unsafe deserialization / dynamic eval Decryption followed by code execution Remove obfuscated code execution patterns. Legitimate code does not need base64-encoded payloads executed via eval, encrypted-then-executed blobs, or dynamic attribute resolution to call system functions. | Manifest | skills/dbhurley/pptx-creator/scripts/use_template.py:147 | |
| MEDIUM | Unsafe deserialization / dynamic eval Decryption followed by code execution Remove obfuscated code execution patterns. Legitimate code does not need base64-encoded payloads executed via eval, encrypted-then-executed blobs, or dynamic attribute resolution to call system functions. | Manifest | skills/dbhurley/pptx-creator/scripts/use_template.py:153 | |
| MEDIUM | Outbound Network Request to Localhost with User-Controlled Input The `create_pptx.py` script initiates an HTTP POST request to `http://localhost:8000/generate_image` with a user-controlled `prompt`. While targeting `localhost` limits direct external exfiltration, this could be used to interact with other sensitive local services, potentially leading to information disclosure, denial of service, or further exploitation if the `generate_image` service is vulnerable to prompt injection or command injection. If the image generation service is intended to be local, ensure it is robustly secured against injection attacks. Consider using a more controlled inter-process communication mechanism or validating/sanitizing the `prompt` string more rigorously before sending it to the external service. If the service is not strictly necessary or can be replaced by an in-process solution, remove the network call. | LLM | scripts/create_pptx.py:300 | |
| LOW | Unpinned Dependencies The skill's Python scripts specify dependencies without pinning them to exact versions (e.g., `python-pptx`, `httpx`, `pyyaml`, `Pillow`). This introduces a supply chain risk, as a malicious update to any of these libraries could compromise the skill without requiring changes to the skill's own code. Pin all dependencies to exact versions (e.g., `python-pptx==0.6.21`). Regularly review and update these pinned versions to incorporate security patches. | LLM | scripts/analyze_template.py:4 |
Scan History
Embed Code
[](https://skillshield.io/report/c35860ae95c477af)
Powered by SkillShield